False Optimism

MSNBC reported in May:

The 2006 Atlantic hurricane season will be very active with up to 10 hurricanes, although not as busy as record-breaking 2005, when Hurricane Katrina and several other monster storms slammed into the United States, the U.S. government’s top climate agency said on Monday.
“NOAA is predicting 13 to 16 named storms, with eight to 10 becoming hurricanes, of which four to six could become ‘major’ hurricanes of Category 3 strength or higher,” said Conrad Lautenbacher, administrator of the National Oceanic and Atmospheric Administration.

But unfortunately for him this never happened:

Instead, it has been a long, lazy hurricane season with just half the number of hurricanes predicted and not a single one making landfall.
"Weather forecasting is a chancy business,” said Hugh Willoughby, a professor of hurricane science at Florida International University. “It’s gotten a lot better, but if you can’t stand being wrong, you shouldn’t be in the business."

Thanks to Tim Blair who reports that

Can’t predict tomorrow’s weather, can’t predict next year’s weather ... yet there’s a scientific consensus about weather in the coming century and beyond. Just as well these folks are adapting to being wrong.


And what about Australian Hurricanes/Cyclones?



This graph on the left as given out by the ABM shows the number and intensity of cyclones from the year 1970 to 1998. They report

that the total number of cyclones has decreased in recent decades. However, the number of stronger cyclones (minimum central pressure less than 970 hPa) appears to have increased slightly.

Interesting. I don’t come to the same conclusion. My statistical analysis of the data given on that graph actually come to the conclusion of a statistically significant decrease in the number of cyclones over the period from 1970 to 1998. (F = 4.99, p = 0.034). For those not in the statistical know-how. Generally scientists prove a significant result if the p value is below 0.05 (5%). This means that there is a 3.4% chance that the decrease in cyclones in this period is due to luck or random variation.

My analysis of stronger cyclones proves no significant increase or decrease over this time period (F = 1.17, p = 0.289). Hence we can conclude, contrary to what was written by the ABM that there is no evidence to prove that strong cyclones have increased or decreased in the past 30 years, and strong evidence that the total number of cycles has significantly decreased.

I guess people see different things in graphs when they want to believe. Lucky we can test this to prove it instead.